git-site-gen/lib/Gsg/MdParse.pm

package Gsg::MdParse;
use strict;
use warnings;
use Log::Log4perl qw(:easy);
use Exporter qw(import);
our @EXPORT_OK = qw(render_readme);

sub link_line($) {

	my $line = shift;
	if ( $line =~ m/(<(http.*)>)/ ) {
		my $markup_link = $1;
		my $link = $2;
		my $link_replace = "<a href=\"$link\">$link</a>";
		$line =~ s/\Q$1\E/$link_replace/g;
		return $line;
	}

	return $line;

}

sub strike_line($) {

	my $line = shift;
	#if ( $line =~ m/~~(.*)~~/ ) {
	if ( $line =~ m/^(.*)~~(.*)~~(.*)/ ) {
		$line = "$1<s>$2</s>$3";
		return $line;
	}

	return $line;

}

# README content is passed in as a var, sub returned an HTML version of the parsed markdown
sub render_readme($$) {

	my $readme_content = shift;
	my $logger = shift;

	my $html_readme;
	# Might be a better way to do this? TODO
	open my $fh, '>>', \$html_readme or die "Can't open variable: $!";

	# Main parsing logic, doing it line by line
	# I have some ideas on how to make this more robust/efficient,
	# but starting with below for POC
	# TODO:
	# Headers parsing can be one concise subroutine
	# Started building a suite of simple subs to handle everything else,
	# each line should be run through the 'sub suite' to handle all parsing
	my @readme_lines = split("\n", $readme_content);
	my $inside_code = 0;
	foreach my $line ( @readme_lines ) {
		# HEADERS
		if ( $line =~ m/^#{1}(?!#)(.*)#$|^#{1}(?!#)(.*)$/ && $inside_code == 0 ) {
			my $parsed_line;
			if ( ! defined $1 || $1 eq "" ) {
				$parsed_line = "<h1>$2</h1>";
			} else {
				$parsed_line = "<h1>$1</h1>";
			}
			print $fh "$parsed_line";
		} elsif ( $line =~ m/^#{2}(?!#)(.*)#{2}$|^#{2}(?!#)(.*)$/ && $inside_code == 0) {
			my $parsed_line;
			if ( ! defined $1 || $1 eq "" ) {
				$parsed_line = "<h2>$2</h2>";
			} else {
				$parsed_line = "<h2>$1</h2>";
			}
			print $fh "$parsed_line";
		} elsif ( $line =~ m/^#{3}(?!#)(.*)#{3}$|^#{3}(?!#)(.*)$/ && $inside_code == 0) {
			my $parsed_line;
			if ( ! defined $1 || $1 eq "" ) {
				$parsed_line = "<h3>$2</h3>";
			} else {
				$parsed_line = "<h3>$1</h3>";
			}
			print $fh "$parsed_line";
		} elsif ( $line =~ m/^#{4}(?!#)(.*)#{4}$|^#{4}(?!#)(.*)$/ && $inside_code == 0) {
			my $parsed_line;
			if ( ! defined $1 || $1 eq "" ) {
				$parsed_line = "<h4>$2</h4>";
			} else {
				$parsed_line = "<h4>$1</h4>";
			}
			print $fh "$parsed_line";
		} elsif ( $line =~ m/^#{5}(?!#)(.*)#{5}$|^#{5}(?!#)(.*)$/ && $inside_code == 0) {
			my $parsed_line;
			if ( ! defined $1 || $1 eq "" ) {
				$parsed_line = "<h5>$2</h5>";
			} else {
				$parsed_line = "<h5>$1</h5>";
			}
			print $fh "$parsed_line";
		} elsif ( $line =~ m/^\*(.*)/ && $inside_code == 0) {
			$line = strike_line($1);
			$line = link_line($line);
			my $parsed_line = "<ul><li>$line</li></ul>";
			print $fh "$parsed_line";
		} elsif ( $line =~ m/^```$/ ) {
			if ( $inside_code == 0 ) {
				$inside_code = 1;
				print $fh "<br>";
			} elsif ( $inside_code == 1 ) {
				$inside_code = 0;
			}
		} elsif ( $inside_code == 1 ) { 
			print $fh "<code>$line</code><br>";
		} elsif ( $line =~ m/(http.*)/ ) {
			$line = link_line($line);
			print $fh "$line<br>";
		} elsif ( $line =~ m/^~~(.*)~~/ ) {
			$line = strike_line($line);
			print $fh "$line<br>";
		}
		else {
			print $fh "$line<br>";
		}
	}

	print $fh "<br>";
	close $fh;
	$logger->info("Parsed README");
	return $html_readme;

}
1;